# Text continuation

A small version of GPT-2 pre-trained on 115GB of cleaned Czech text, suitable for Czech text generation tasks

Chinese language model developed based on bigscience/bloom-1b7 architecture with 1.4B parameters, reducing GPU memory usage through vocabulary compression

This is a Spanish GPT-2 model trained from scratch on the large_spanish_corpus (BETO corpus) using the Flax framework, developed with support from the HuggingFace community week event.

This is a small-scale Spanish T5 model trained on the BETO corpus using the Flax framework, supported by the HuggingFace community week event

This is a fine-tuned version of the GPT2-Small English model, trained on Turkish Wikipedia articles, suitable for Turkish text generation tasks.

ParsGPT2 is a Persian version of the GPT-2 model, developed by the Hooshvare team for Persian text generation tasks.